New skin test for detection of bovine tuberculosis on the basis of antigen-displaying polyester inclusions produced by recombinant Escherichia coli

Abstract

The tuberculin skin test for diagnosing tuberculosis (TB) in cattle lacks specificity if animals are sensitized to environmental mycobacteria, as some antigens in purified protein derivative (PPD) prepared from Mycobacterium bovis are present in nonpathogenic mycobacteria. Three immunodominant TB antigens, ESAT6, CFP10, and Rv3615c, are present in members of the pathogenic Mycobacterium tuberculosis complex but absent from the majority of environmental mycobacteria. These TB antigens have the potential to enhance skin test specificity. To increase their immunogenicity, these antigens were displayed on polyester beads by translationally fusing them to a polyhydroxyalkanoate (PHA) synthase which mediated formation of antigen-displaying inclusions in recombinant Escherichia coli. The most common form of these inclusions is poly(3-hydroxybutyric acid) (PHB). The respective fusion proteins displayed on these PHB inclusions (beads) were identified using tryptic peptide fingerprinting analysis in combination with matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). The surface exposure and accessibility of antigens were assessed by enzyme-linked immunosorbent assay (ELISA). Polyester beads displaying all three TB antigens showed greater reactivity with TB antigen-specific antibody than did beads displaying only one TB antigen. This was neither due to cross-reactivity of antibodies with the other two antigens nor due to differences in protein expression levels between beads displaying single or three TB antigens. The triple-antigen-displaying polyester beads were used for skin testing of cattle and detected all cattle experimentally infected with M. bovis with no false-positive reactions observed in those sensitized to environmental mycobacteria. The results suggested applicability of TB antigen-displaying polyester inclusions as diagnostic reagents for distinguishing TB-infected from noninfected animals.

FIG 1

Schematic representation of hybrid genes encoding fusion proteins which mediated production of polyester beads displaying TB antigens.

FIG 2

TEM analysis of recombinant E. coli harboring various plasmids (A to E) and of isolated polyester beads displaying mycobacterial antigens (F to J). (A and F) pET-14b phaC; (B and G) pET-14b cfp10-phaC; (C and H) pET-14b rv3615c-phaC; (D and I) pET-14b esat6-phaC; (E and J) pET-14b cfp10-rv3615c-phaC-esat6.

FIG 3

Protein profiles of polyester beads isolated from recombinant E. coli harboring various plasmids without (A) and with (B) a protease inhibitor treatment. (A) Samples without protease inhibitor treatment during bead extraction. Lane 1, molecular weight marker (Mark 12; Invitrogen); lane 2, PhaC (wild type, 63 kDa); lane 3, Rv3615c-PhaC (75.2 kDa); lane 4, ESAT6-PhaC (74.3 kDa); lane 5, CFP10-PhaC (75.2 kDa); lane 6, CFP10-Rv3615c-PhaC-ESAT6 (98.1 kDa). (B) Samples with protease inhibitor treatment during bead isolation. Lane 1, molecular weight marker; lane 2, PhaC; lane 3, CFP10-PhaC; lane 4, CFP10-Rv3615c-PhaC-ESAT6. The presence of PhaC-TB antigen fusion proteins was confirmed by tryptic peptide fingerprinting using MALDI-TOF MS (see Table S2 in the supplemental material).

FIG 4

Model of a triple-TB-antigen-displaying polyester bead produced by recombinant E. coli.

FIG 5

Assessment of TB antigen reactivity in vitro. The ELISA is described in Materials and Methods. All measurements were conducted in triplicate. The mean of TB antigen reactivity to a specific antibody is reported ± the standard deviation. (A) Reactivity of TB antigens on the surface of polyester beads. Wild type (WT) (beads) is the negative control. Recombinant TB antigens (CFP10, Rv3615c, ESAT6, and the mixture of CFP10, Rv3615c, and ESAT6) were used as positive controls. The testing samples are immobilized TB antigens (bead-CFP10, bead-Rv3615c, bead-ESAT6, and bead-CFP10-Rv3615c-ESAT6). The anti-CFP10 and anti-ESAT6 were mouse monoclonal antibodies, and the anti-RV3615C polyclonal antibody was produced in a rabbit. *, significantly higher than single antigen (P < 0.05) (B) Control experiment; the rabbit preimmune serum was used to replace the three specific antibodies as indicated in panel A. (C) Cross-reactivities of specific antibodies with the other two TB antigens. The total protein concentrations of control beads, TB antigen beads, and free recombinant antigens in this figure were 1 μg/ml, 1 μg/ml, and 0.125 μg/ml, respectively. *, significantly higher than all other beads (P < 0.05).

FIG 6

TB skin test response. Four diagnostic reagents, avian PPD, bovine PPD, beads displaying three TB antigens (Ags), and control beads, were used to skin test cattle. The result was positive if the increase in skin thickness was ≥1 mm. (A) Cattle experimentally infected by M. bovis. (B) Noninfected cattle naturally exposed to environmental mycobacteria. The bar indicates the median.